Chapter 1
[pic]
Chapter 2
Different Paths to Clinical SAS Programming
Associate Director of Bioanalyst – David Tappe 2
Many Roads Traveled
There are many paths that individuals take to arrive at a career as a SAS Programmer. SAS is a programming language which requires the rigors of a computer scientist. However, to become a clinical trials analyst also requires a thorough understanding and knowledge of the clinical data to be analyzed. The combination of these distinct skills attracts people from diverse backgrounds. It is surprising to find so many different backgrounds of individuals with differing talents and interests who gravitate towards this career path. Like many other career paths dealing with technology, things change. The dynamic aspects of the job require programmers to become resourceful in order to survive. They therefore must re-invent themselves continuously. The core understanding of the analytics behind clinical trials is still evolving but remains comparatively stable. Once you gain a thorough understanding of clinical data and the accompanying programming techniques, you then have a good foundation for becoming a successful Clinical Trials Programmer.
This chapter will evaluate several different paths that individuals took on their road through clinical SAS programming. There is no one correct path but rather an abundance and diverse set. The diversity of people who succeed in this field illustrates the fact that it requires a unique combination of skill sets needed to thrive in this environment. This career path has only recently been established so many of the individuals who entered this field did not initially set out to become a clinical SAS programmer. For many, the journey was unplanned and there were no road maps. They are pioneers and navigated their career paths following their intuition and relied on luck. There are many lessons that can be gained from these pioneers. The paths which these individuals paved will provide a clearer vision and set of directions for future aspiring SAS programmers who plan to venture on the same road.
Associate Director of Bioanalysis – David Tappe
It is not about how good a SAS programmer you are. It is about being focused on the data and your representation of the data that leads to success.
Starting Out
When David started out on his first SAS programming position at Syntex in Palo Alto, California in the 80s, a clinical trial analyst and the associated work was not well understood nor well known on the west coast. The term “Bioanalyst” used at Syntex for statistical programmers was foreign and he did not know what to expect.
David went to UCLA to obtain his bachelor’s degree in Mathematics. He then continued his studies at Fresno State in Biology and did his thesis work in Lymnology, or the study of fresh water ecosystems. His academic experience did not include formal computer science nor instructed him on how to program in SAS. It did however provided exposure to research methods and statistics that he used in his thesis. In fact, David had little formal programming training. He was introduced to programming when he took a FORTRAN course at UCLA. Back in the days of punch cards, he used FORTRAN for an ecosystem modeling class in graduate school. Out of personal interest, he learned BASIC programming on his own while working as a Lab Technician at U.C. Berkeley. As he developed his career, the technical aspects of programming were merely a means to an end. That end was the data itself and how the data demonstrated meaning when analyzed. The meaning of the data was therefore pivotal and overrode the mechanics of getting there. David gained enough proficiency in programming on the job to perform well without requiring much formal training.
In 1979, David moved to Walnut Creek, which is east of the San Francisco bay, after he had completed his course work in Fresno. He explored jobs in the bay area and first worked in the department of Materials Science at U.C. Berkeley helping on research projects aimed at finding ways to utilize coal dust. As that job was ending, he found a job announcement at the U.C. Berkeley placement center for a Bioanlyst at Syntex. Syntex was an established pharmaceutical company, but due to the lack of experienced statistical programmers in the bay area at the time, they hired people at entry level. Although he lacked the job specific experience, he applied for the job at Syntex because he was intrigued by the blend of statistics, health research and programming mentioned in the job announcement. He did not know what a Bioanalyst did or what the job entailed. The commute from Walnut Creek to Palo Alto required him to go over the east bay hills, across one of the bay bridges. This added to his commute totaling about 60 miles. Even though the commute was brutal, the job proved to be a good fit and kept him coming back. In the end, it was the research aspect and ability to understand the meaning behind that data that cemented his decision to stay in this field as a career.
He began to appreciate the goal of producing study reports summarizing the safety and efficacy results of clinical trials. While he enjoyed programming and developing good SAS programming skills, the output produced by his programs were more important than the SAS program itself. He began to realize that this was the essential aspect of the Bioanalyst’s job. The Biostatistician, who was in a way the Bioanalyst’s main client, wanted to receive quality output tables representing the study data and its analyses results. They were not interested in the quality of the SAS program used to create the tables, so long as they produced accurate results.
The intricacies of disease that David was navigating in his job was a constant source of fascination. He is not clinically trained as a MD or a RN, but his interest in research and learning about the disease related to a given drug allowed him a unique view of the data. This was a window through the data which provided meaning that became essential in his work. The interactions with other team members mostly involved communicating with the Biostatistician. However, it also included interacting with the other clinical team members, such as the MD in charge of the project. The Biostatistician and Bioanalyst worked together closely in analyzing and producing report tables and graphs. Even though the meaning of the statistics used is in the domain of the statistician, it often ended up with the Bioanalyst needing to interpret their instructions to fill in the gaps. This multi-faceted aspect of the job kept David intrigued and sustained his long commute for 12 years.
Growing within the Job
Although the data kept David interested, he did continue to explore technical aspects of the job as well. He continued to develop his SAS programming skills by taking some database design courses at U.C. Berkeley Extension. This gave him the opportunity to work on several Oracle database projects. He also was part of the initial Rx/Clinical development team, a joint venture between Oracle and Syntex that led to the product now known as Oracle Clinical. This became the leading clinical trial database management system used in the pharmaceutical industry.
He began to take upon management responsibilities, managing both people and projects at Syntex. This continued on in his career at Chiron in Emeryville. He modestly attributed the management of people as being an easy job because the people he worked with were capable and good to work with. Since he had a good understanding of what it took to perform as a Bioanalyst, he was able to identify people who had the right balance of research and interest in the data, along with the technical programming skills. It was a combination of having the tenacity to acquire an understanding of the clinical aspects of the data, while also having the technical skills to perform analysis. This was what David had strived for himself and was what he looked for in individuals as he formed the teams which he managed.
David is now the Associate Director of Bioanalysis at Novartis (formerly Chiron). His first goal at Chiron was to establish a standard set of reporting macros which would handle the standard CRF (Case Report Forms). He decided to have his Bioanalysts balance their workload of producing the day-to-day analysis and reporting work with developing and maintaining a standard library of macro software. The standard software mostly focused on generating safety summary reports since the efficacy data usually varied too much between projects. The difference among studies didn’t lend itself to standardization. He proceeded to assign Phil to work on Lab data, Frances to develop macros for Adverse Events, Rob to work on Demographics and Conmed while David himself worked on the Dosing information. It was a management style that allowed him to participate in the work firsthand, while motivating the team to develop software together. This was all implemented at the same time as the normal study analysis work which they had to do. He enjoyed working on these macros which organized dosing information into a dosing analysis file. This allowed integrating a subject’s dosing experience with other study data such as adverse event and efficacy data. For example, an adverse event can occur at any specific time during the study. This centralized data mart approach is a useful technique in managing clinical data. The macros being developed would then merge this data with the dosing information and added variables such as the exact dose, the cumulative dose, time since last dose, etc... The merged data would allow for the analysis to infer and answer important key safety questions that the study protocol needed to answer.
David’s approach to developing the macro system was to standardize any area where the potential re-use outweighed the development and validation costs. Having a standard macro system available provided for consistent analyses across studies, projects and output. It also prevented the programmer from having to do repetitive tasks. This allowed them to focus on the unique efficacy aspects of each study. It made the job easier and more efficient for the Bioanalyst. His team was also more engaged and interested since the work was more dynamic and became more like a unique puzzle that needed to be solved for every study.
It took a year to develop the first version of the software with much of the team’s efforts focused on the macro system without a lot of competing study analysis work. Later on, it took more time to implement subsequent version releases since the programmers usually had traditional project analysis work that took precedence over the standard macro development. David hypothesizes however that this was a more cost effective approach. By giving the development efforts to the senior programmers who work with the analyses themselves, the development of software was implemented by the same individuals who understand it the most.
David uses a low key approach to managing his people. He sets out goals and expects his people to work conscientiously and autonomously. He does not see himself as having a draconian style but rather a gentle coaching approach. David performs peer reviews and makes suggestions but, in general, leaves his people alone. He sees that this approach provides the freedom that fosters the talents of each individual to thrive and discover their own unique balance of various skills to be a successful Bioanalyst.
Consulting SAS Programmer – Gene
Being resourceful and learning new things will keep you from becoming stagnant and help you become successful as a SAS programmer.
Back in the Day
Gene’s introduction to computers and technology started back in high school in the 70’s when he used the Texas Instrument calculators that allowed him to program equations. He then learned assembly language and machine code on DEC PDP-11 and Zilog Z-80 microprocessors. Gene enjoyed working at the machine level while using these 16 and 8-bit machines because it taught him the fundamentals and rigorous approach towards programming and debugging programs. He then progressed into learning the BASIC programming language. In addition to Gene’s disposition towards problem solving which attracted him towards computers, he also had an interest in becoming a “doctor” so he started his undergraduate work at Johns Hopkins in 1980 on a degree in Biomedical Engineering. This combined his clinical aspirations with his analytical, research-oriented and engineering ambitions.
The university setting, along with summer jobs, gave Gene experience working on research projects. Even though he spent a lot of time in the lab working on animals, technology was always present. For example, the research labs had a Radio Shack TRS-80 computer controlling a dog’s heart rate. The combination of technology as applied to medical research was fascinating because it was very “Frankenstein”. Similar to how tinkering with early 8-bit computers gave Gene the fundamentals of software development, the laboratory gave him a foundation for medical research. This environment aided him in his work on clinical trials.
Although he did not use SAS yet, Johns Hopkins and his graduate work at Case Western Reserve University did allow Gene to work with BMPD. This was another statistical software package at the time which allowed him to run statistical models such as ANOVA and T-TEST which were in routines similar to SAS PROCs.
The First Job
One of the first jobs that Gene started to work on after his Masters degree in Biomedical Engineering was working as a Data Manager in Los Angeles for a non-profit organization named Los Angeles Regional Family Planning Council. He worked on projects dealing with clinical trials on contraceptives products such as the IUD and non-latex condoms. Gene worked closely with the data functioning as a DBA/Data Manager using dBase III with tasks such as defining the database structure and setting up data entry screens. This is where he was first introduced to SAS but had limited knowledge of SAS since there were no other SAS programmers in the entire organization. He therefore referred to his manager who had a statistical background, but also had limited experience in programming SAS. Gene enjoyed learning programming on his own and purchased additional SAS manuals directly from SAS Institute to teach himself how to program in SAS. .
As a Database Manager, Gene worked closely on data discrepancies such as programming edit checks and the process of transferring data from dBase to SAS via flat ASCII files. Once he had the data in SAS, he was able to generate descriptive statistics using PROC MEANS, FREQ and UNIVARIATE. The programming practices being conducted were not as formally structured as in a typical clinical trial. As a general rule, permanent analysis files were never saved, but instead the same SAS program would import the source data, create temporary analysis files, and then directly generate the final statistical reports. At this point, Gene only dabbled in SAS and it was not yet his main job.
After leaving LA, he moved to the San Francisco Bay Area to work for a medical device company dealing with laser eye surgery named VISX. He sought out this new position as a biomedical engineer, since it matched his college education. Although he did initially accomplish some engineering improvements on their product, his background in working with clinical data ultimately dictated his role at VISX. Since Gene had been working as a data manager, they had placed him in a position working closely with their DOS-based Paradox database with analysis in SAS. He continued working on the data discrepancy management which was then exported to SAS for analysis. Since this medical device was VISX’s major product and its major competitor, Summit Technologies, was developing a similar product; there were tremendous timeline pressures to get the data submitted to the FDA for approval to defeat their competitor on the open market. VISX therefore decided to hire SAS consultants to assist in achieving this goal. Although marketing approval for the VISX’s product was virtually completed in Canada and Europe, they were still in the process of getting approval in the USA and therefore were rushing to submit the clinical trial reports to the FDA. Part of Gene’s responsibilities was to guide the SAS consultants in their jobs. He was not their formal manager but he worked closely with them since he understood the meaning of the data and had some experience with SAS programming. He could bridge the gap between what the statisticians wanted to report on and how SAS programming could accomplish these tasks. This was an eye opening experience in that for the first time, he discovered that there existed such a job where you can program in SAS 100% of the time. This intrigued Gene since he did not care for the administrative aspect of people managing which was part of his responsibility. These SAS consultants would just come in, focus on their work and were not involved in as much meetings or office politics. These guys were “do-ers” not “talkers”. They did not have administrative tasks such as dealing with yearly performance reviews and were just there to do a specific job. Gene saw this as a refreshing alternative way of working. He also liked how these consultants had opportunities to experience how different companies operated since they did not stay at one company as long as permanent employees. This variety of work kept these consultants on their toes and prevented them from falling into a rut of routine tasks that were boring and stagnating.
This insight encouraged Gene to continue to learn more SAS by interacting more closely with the consultants. He also learned by seeing sample code that these consultants would share with him. Gene was now spending 50% of his time working on SAS which was more than he had ever done before.
Transition to Consulting
There were corporate challenges at VISX which lead to layoffs. This and other management turmoil lead Gene to try out the consulting world starting with work at Roche Biosciences. He started out with a three month contract which is typical for an initial contract. This introductory period allowed the company to evaluate the consultant to see if it worked out without making a big commitment. If things went smoothly and there was work to be done, the contract normally got extended. Gene worked on the precursor to Oracle Clinical known as Rx/Clinical which is the clinical trials relational database jointly developed by Oracle and Syntex. Syntex also later became Roche Biosciences. Even though he started out working in the role as a data manager, Gene grew into the position of a SAS programmer and continued to work as a consultant at Roche for three and a half years. This was where he learned the majority of his SAS programming skills. During these same years, Gene noticed that there were some individuals in this large organization who began to have a “DMV” mentality. After years of working in a large bureaucratic organization, they lacked the desire to learn new things and became less motivated. Some of his colleagues were in the same company for tens of years. He began to see how this stagnation crept into his work routine as well, even though he was a consultant. This realization woke Gene up to the reality that perhaps he was also falling into this stagnation. In attempts to being true to his original motivation for becoming a consultant, he decided to make a change.
Gene ventured into a short contract with Quintiles Pacific down the Peninsula in Mountain View. This was the first time he worked within a Clinical Research Organization (CRO) which was different from his experience with a pharmaceutical company. His initial thoughts were that since it was a service organization, the work would be varied since they were servicing many different clients at the same time. It turned out however that many clients rely on the CRO to dictate the standards. It therefore turned out that the work was still done mostly in the same way across different clients and projects.
The desire to learn new things and to work in different environments drove Gene to search for new projects. He had heard from other SAS consultants that finding jobs on your own was the way to earn the biggest salary. Despite this fact, Gene relied on job placement firms such as Trilogy and ASG to land him new opportunities. Gene preferred working in this manner because he preferred to take advantage of the marketing resources of these firms to find him the projects so he could focus on “doing the work and not trying to find the work”. With a little luck, there is always something new and interesting to learn within each work environment. At Pharmacyclics, he worked with the Trialex System which was a browser based thin client software to manage reports and analysis files he created using SAS. It was a dynamic time with new tools and methodologies to keep him on his toes.
Now living in North Carolina, he works for Schwarz Bioscience which is based in Germany. In this position, Gene works at home most of the time connecting to servers in Germany. When Gene started, they did not have cube space in their North Carolina office, so the arrangement was for him to connect via his cable modem from home and work through a remote desktop using Citrix remote desktop technologies. Most of the time, it worked well but once in a while, his keyboard would spontaneously switch to a German style keyboard and his Ultraedit editor would create unexpected characters so he would have to log out and back in to regain the use of an English keyboard.
Different organizations have different approaches towards validation of SAS programs. This is partly due to varying interpretations of the regulations and also due to how different managers and organizations function. At this current project, Schwarz Biosciences has an approach of validating all the output 100%. This means that there is an independent programmer who would re-produce the same numbers in a separate SAS program stored in a separate SAS dataset which is distinct from those that are meant to be submitted to the FDA. The new dataset would then be compared to the submitted data through the use of PROC COMPARE to ensure that the two sets of numbers are identical before it is deemed “validated”. When Gene first started, he primarily did the duplicate validation type programming. Now that he has been on the project longer, he is appointed to perform more of the original programming that will be part of the submission.
Since he works remotely, Gene’s main method of communicating with his coworkers is through email. Ninety percent of the time, his email and communications are dealing with detailed information such as data, file, and variable names. He does recognize that at times, it does take a little longer to describe or show someone something in an email compared to just being physically in front of them. Even with these challenges, Gene has adapted and rather prefers this arrangement and feels perhaps that this is the direction in which SAS programming is going to move towards in the future.
Resourceful Reinvention
Gene has found many creative and resourceful ways to learn and reinvent himself as a SAS consultant. Even before he started working professionally, his interest in 8-bit microprocessors and working at the electronic circuitry level coincided with his tinkering tendencies and attention to detail. His personality gravitated towards working with the intricacies of data and managing its discrepancies. If Gene were to find one data issue, it would bother him to just leave the rest alone. Similar to an irritating itch, he had to then investigate the rest of the data or system to ensure that there are no such problems occurring elsewhere. There is a certain satisfaction and comfort knowing that he can ferret out all the discrepancies that exist in a database. An inclination to focus on details also helps in debugging SAS programs created by others.
He did not attend SAS training courses. Instead, Gene was able to get his consulting firm to pay for the SAS reference manuals he requested. User manuals proved to be a more cost effective way of learning and became a continuous source of knowledge that can be referenced time and time again. Good reference manuals had sample code that could be used as templates to algorithms in real world applications. Gene started his own personal electronic library of sample code that he refers to on occasion. This process started in his early days when he first started working with other SAS consultants prior to his becoming a full time SAS programmer. Since then, he continued adding to it including original algorithms he came up with and code that he acquired while attending SAS conferences. The conferences are a resource that can not be ignored. Gene found that SUGI, which has evolved to what is now known as SAS Global Forum, is the biggest conference and also the most expensive. There seems to be a larger representation of seasoned SAS programmers and managers that attend SUGI as compared to other SAS conferences. He perceives SUGI as not necessarily focused on the Pharmaceutical industry, so it is only by chance that he would find a paper with substance and relevance. Gene therefore prefers the smaller regional conferences or PharmaSUG where there is more diversity of attendees and focus on talks that are more relevant for him. Conference and user manuals are just examples of a few resources which Gene taps into for gathering new ideas to build his own set of SAS knowledge. Gene is constantly sharpening his skills and looking for insight by these many resourceful approaches. It is a Darwinian world where the one that sharpens his tools and constantly reinvents himself by learning new thingsis the one that survives and thrives.
Manager of Statistical Programmers – Patty Gerend
Combining basic research with interest in computer programming can result in a rewarding job as a SAS programmer.
The Need for SAS
Patty had never set out to program in SAS when she was studying biology and biochemistry at the University of Connecticut and continuing on to her Masters degree at West Leyman. She had never even heard of SAS until it was first mentioned at her interview for her job at Syntex in 1985. She did not really take any computer programming classes in school since she was more focused on the area of physical science within her curriculum.
Patty started to work as a biochemist when she got out of school at UCSF doing research on diabetics and other related studies on the effects of zero gravity. Although this was different from programming in SAS, the methodologies of research and its relationship to clinical data was the common thread that would later lead Patty to SAS. Patty spent many hours working in the lab with chemicals and animals while compiling related research data. The data was structured differently from clinical trials but it required a thorough understanding prior to doing data analysis. In the design of the experiment which Patty worked on, there were no formal analysis plans but Patty took a large role in doing everything herself from deciding what information to collect, to capturing the information and then performing analysis on this. This was in the late seventies and early eighties so it predated MS Excel. Patty therefore used HP programmable calculators and Word Start to type up her findings. She later started to use Lotus 123 to perform her linear regression analysis. This was a challenge since there were no pre-built routines such as SAS PROCs so Patty had to program these algorithms herself in order to get the statistics that she needed.
After working in this academic research environment for a few years, Patty realized that the work was based on grants and had limited resources. There were no opportunities for advancement and she did not feel she could grow. The need to constantly seek grants made her feel that her job was at risk and not very stable. She then began to explore different career paths and started to take computer classes at San Francisco State. Her studies included BASIC, FORTRAN and Pascal. These courses were a good introduction, providing her an appreciation and interest in computer science and programming for the first time. Although her interest was growing in her night classes, Patty also continued some of her classes in her field of work. It was a pivotal moment in her recombinant DNA certificate class when she realized that it was time for a change in career to pursue a job that allowed her to program on a computer.
Career Change
Patty landed a position in 1985 as a Bioanalyst at Syntex which was an entry level position. Her supervisor Marty was looking for someone with research background yet also had computer skills. The night classes which Patty took paid off since they were a good combination with her research background to secure her first job in this career change. Syntex was a good environment as an introduction to becoming a clinical SAS programmer since it was a large organization with many resources. Patty learned different aspects of clinical trials from her interactions with statisticians including her boss who was also a statistician. She also worked with data managers who gave her more in depth knowledge of the process of capturing structured clinical data. She then joined a group within Syntex which worked on a Unified Corporate Safety Database (UCSD) supporting many FDA filings dealing primarily with safety data. At the time, SAS was running within a batch environment on an AMDAL mainframe. Patty had used JCL to submit her batch jobs which took data from older databases such as IDMS and transformed them into the standard structures of UCSD. This enabled Patty to learn more about SAS data step and its data transformation capabilities.
It was during this time that Roche acquired Syntex and the research division that Patty was working in was being downsized. Patty ventured into SAS consultancy for a short time. She did not like the aspect of having to sell and market herself. The idea of worrying and not knowing when the next project would come along was something that she did not enjoy so this was not a good fit for her.
Managing Programmers
She then joined Genentech and started to grow more in the area of becoming a manager of SAS programmers. Genentech had a new structure working with study teams that Patty enjoyed working in. As she began to take upon more responsibilities in managing, she did less SAS programming. She recognized that there were two aspects to her work. One area dealt more with the “industry specific” type of work and the other was “business administration”. The industry specific work included tasks such as designing studies and developing analysis on the clinical trials data. The business administrative work dealt more with budgeting and cost modeling for the team. Patty enjoyed this balance and was spending about half of her time doing each. The work load combined tasks that required people skills and gaining an understanding of the many facets of management including things such as legal responsibility in working with people and how to optimize their performance. On the other hand, she also learned industry specific tasks dealing with electronic submission that is still being developed and therefore proved to be dynamically interesting.
Genentech had the resources to send Patty to many courses to support her growth and interest as a manager. Some training materials were online while others were instructor based. In addition to management classes, she also took advantage of the technical classes on SAS macros and on relational databases using Oracle. Besides taking advantages of the many resources at Genentech, Patty also learned more about the industry specific tasks at conferences such as DIA (Drug Information Association).
Patty has a good balance of analytical skills which provide her the aptitude to excel in the technical aspects of SAS programming combined with an even keeled temperament and personality to work with a diverse group of people. Patty draws an analogy of her team with a biological model. Like any population within a species in biology, diversity is a key component of survival. If the individuals within the group all have the same qualities, it is prone to die off in the event of a disease or physical adversity. This biodiversity can also be applied to skills and personalities of people within a team of SAS programmers. Patty tries to foster diversity in the team she works with since the various talents that people bring ensures its success in dealing with many unexpected twists and turns of projects. Patty sees many different types of personalities in her group ranging from ambitious visionaries to low key getting the job done types. She sees herself being somewhere in the middle yet having the appreciation for the range. This allows her to connect to each team member resulting in a unique and effective management style.
Manager of Statistical Programmers – Deena Gruver
SAS programming is just one aspect of a larger process of clinical research. An appreciation of all the tasks as a whole provides a greater understanding of the job.
Introduction to SAS
Deena started using SAS to do some of her assignments at Bowling Green State University in Ohio during the late 80s. She had used it to apply some of the statistical models for her degree in applied statistics. She programmed on a VAX system and had most of her results printed on a dot matrix green bannered spooled paper. At the time, she did not appreciate the programming aspects of SAS and had used it mainly as a means to get the statistical results. It was just a tool to get the answers she needed for her course work. It therefore never caught her interest.
After school, Deena started a job with Trilogy at Kalamazoo. Trilogy was a consulting firm and its main customer was Upjohn which was just down the road in Kalamazoo. Although she was initially hired on as a Statistician, since she had just completed her masters in this area, Trilogy had her doing mostly SAS programming. The work environment was different from all that she was used to in her previous academic experience because Trilogy had hired about 40 new employees who all seemed to be new to SAS and clinical trials. She met her supervisor on a weekly basis to go over projects but she relied on the online training that Trilogy provided along with assistance from fellow co-workers to learn SAS. It felt like everyone was in the same boat asking each other for help and comparing notes on approaches to each programming task. The deliverables to Upjohn were summary tables, data listings and figures. The analysis files were mostly already done so many of the jobs focused on a narrow process of producing the reports.
Deena continued to work with Trilogy as they placed her at Lifescan, which was a medical device company in Milpitas, California. This was slightly different from her six month stint at Kalamazoo in that she was working with the employees of the sponsor company on a day-to-day basis rather than with other fellow Trilogy consultants. She learned a lot from interacting with other senior level SAS programmers. What was a pivotal moment for Deena was taking a class in SAS macro language. This gave her the tool to parameterize some of the programming routines. It therefore allowed her to modularize her programming and made repetitive tasks much more efficient. Although she did learn a lot from her experience with Trilogy, she still did not feel she had a complete understanding of her job since as a consultant; she was assigned specific jobs and therefore did not work on the entire project as a whole.
Transition to Permanent
With this dissatisfaction, Deena left Trilogy and started a SAS programming job at Kera Vision as a permanent employee. As a permanent employee, she was now able to see the whole process of what it takes to put together a PMA. She felt now that she belonged to the team and there was a sense of ownership, that she had a stake in the project and company. She had many curiosities about the bigger pictures and now it was being revealed to her. She was also able to interact with a more diverse staff in this new position. Besides the statisticians or other SAS programmers, Deena talked to CRAs and other members of the clinical staff along with data management. She took requests from CRAs and went to the source of the data with the data management group to resolve the issues. The whole picture of the clinical trials process was beginning to be clearer now and she felt a greater sense of ownership along with the stability in this new position. It was good not to have to worry about what was going to happen next. There was some restructuring at Kera Vision so Deena had an opportunity to become a manager. She enjoyed the extra responsibilities but then the company began to downsize at the end of 2000.
Deena then took a new position at Pharmacyclics as a Senior SAS programmer. The work environment at Pharmacyclics paired SAS programmers with biostatisticians and they worked closely as a team. Deena enjoyed the experience of learning to work collaboratively with Jennifer who was the statistician. Deena was disappointed, however, that they were in a separate building from the CRA and other staff members. She felt more isolated here as compared to her previous experiences.
Don, who was her previous boss at Kera Vision, was forming a new team at Protein Design Labs in Fremont and had asked her to join. This time, it was not as a SAS programmer but as a manager of SAS programmers. Deena had some experience in this but this provided an opportunity to start things from the beginning. Transitioning from a programmer into a management position was challenging at first. Deena did not want to delegate her work but had to do so since her role was no longer to do the programming herself. She now only programmed SAS about five percent of the time. She was more focused on other aspects of managing the group of programmers such as dealing with time lines and resource negotiations.
Holistic Approach
Upon reflection, what drew Deena into SAS programming was her aptitude for logical thinking, problem solving and the ability to organize her work. SAS was a big puzzle that she had to figure out. Some of the same skills did lend itself to management as well but managing people complemented other aspects of Deena’s personality. She is now able to combine her ability to connect with people with the technical skills she had learned. Even though she enjoys management, Deena does feel left out at times when her programmers learn new things such as ODS and she does not have the time to try out that new feature. In general, she does enjoy this new role and she is still able to learn a great deal from Don who has become both her manager and mentor. In addition to learning on the job, Deena also tries to grow by going to conferences. Although she enjoys SAS conferences, DIA remains her favorite. DIA is different from SAS conferences in that it covers many aspects of the clinical trials process outside of SAS programming. It covers areas such as regulatory which really helps her understand the bigger picture and become a better manager. Being a manager of SAS programmers seems to fit her better in that it does provide her with a holistic view so she understands the bigger picture of her job.
Telecommute Programmer – Angela Ringelberg
Communicating effectively while adapting to changes are key attributes to becoming a successful telecommuting SAS programmer.
Angela earned a computer science degree in 1986 at the University of Mobile Alabama. She did not program in SAS during school, but did use PL1 as the primary programming language. She also did assembler programming and did not apply it towards statistical models such as in SAS. When she graduated, it was difficult to get a job since she was caught in the bind of not having the necessary prerequisite experience. Angela had wanted to do something related to her technical training in computer science so she started to work through temporary placement agencies.
Angela started out as a data entry clerk for McDonald Douglas in Phoenix, Arizona. She continued to be shuffled around on various temporary projects doing data entry. These jobs did not fit within any specific career path or industry of her aspiration since the placements were more entry level. She then started to work through Jean Simpson, a local small agency in Shreveport, LA, which placed her at Boots Pharmaceutical. It was another data entry position but after three months on the job, she progressed to become a SAS programmer. She was able to accomplish this through talking to various people in the organization and, in particular, the biostatistics and data management departments. Her first introduction to SAS programming was doing edit checks. This was related to what she was doing in that it included algorithms which checked for the accuracy of the data which she worked with during her data entry position. Although she did not work on analyzing the data statistically, the data management tasks in conjunction with using the Clintrials database, gave her a better understanding of clinical trials data. Her programming tasks expanded into programming ad hoc requests for patient listings and data listings. Most of the knowledge was obtained on the job but she did also take some classes from SAS Institute which taught her more about SAS macros and SAS/Graph. Angela stayed at Boots for five years and worked her way up to becoming a supervisor.
Her husband then received a transfer in his job to Southern California at Edwards Air Force Base. It was about two hours away from Los Angeles but it was a smaller and more isolated community. The closest biotech company was Amgen but that would have been a two and a half hour commute, so that was out of the question. Even though she was in California, Angela received a request from her former employer at Boots Pharmaceutical to work on a CANDA project. At the time, many companies were preparing a computer aided new drug application so that the FDA could review the information through the computer. It was a boon for programmers because many companies needed this solution. The problem was that each company had their own CANDA format which made it challenging for FDA reviewers. There was a large learning curve for reviewers to use CANDAs between different companies. As a result, this method is no longer used but it did present an opportunity for Angela to work remotely at that time. She was able to use her own PC to log into the remote machine. The employer even provided her with access to the SAS software on her local PC to do additional development.
Angela began to explore other telecommuting projects through the placement firm NovaTech. She worked on projects for Knoll Pharmaceutical, Quintiles, Norvatis, and Norfolk Railroad among others. It worked well since she was able to do this job at her home. She began to master the art of telecommuting. Angela was then known as the queen of email. This was her main form of communication in addition to the phone. She worked for Novatech for six years. Near the end, she picked up a project for a CRO named PharmaNet. She was a contractor at first working remotely since Pharmanet is headquartered in New Jersey. They had delivered to her a machine and a printer with all the security and software pre-installed. With a DSL connection, she was able to log on via the VPN and she was able to do work from then on. There were some challenges at first since she had to sometimes wake up early to attend 6am meetings Pacific Time which was 9am on the east coast. She primarily communicated with the project manager and statistician remotely. All the annotated CRF were scanned electronically and emailed to Angela. If she needed to see it on paper, she could just print it out. She continued to do some validation programming but the majority of the time was spent developing summary tables, listings, and graphs.
Most of the meetings were done over the phone. However, Angela did also experiment with Net Meeting and MShow which allowed desktop sharing. Even though she started as a consultant, within a period of five months, PharmaNet hired her on as a permanent employee.
Angela has a knack for working as a telecommuter. She feels that one of the primary skills needed is to have excellent communication skills. This can make or break the job since everything has to be communicated though the phone or email. She also has the passion for doing this since it gave her better control over her schedule and better quality of life by being closer to her kids. This also requires self-discipline since she has to construct her own self imposed 8 hour work days without supervisor enforcement. One skill she picked up was to keep tabs on her projects and communicate her status to various team members constantly. That way, they would know where she is at and send her the right amount of projects at the right time. Among these skills sets, she also had to gain a certain level of trust among her co-workers and supervisor. This was established at the beginning and worked out under the initial contract. During that trial period, if things didn’t work out, PharmaNet could easily end the contract without any legal difficulties.
Angela has a good combination of communication skills along with being very adaptable in her work. She made the pivotal move from the data entry clerk into SAS programming by communicating effectively to the right team. She then transitioned and adapted to the various programming requests that came her way even though she only had experience in data entry. When she was relocated to California to a small community, this outpost did not deter her from doing her work. She was able to adapt and re-invent herself as a telecommuting programmer. The technology in telecommunications combined with her talents enables Angela to be a productive and valuable resource no matter where she is located.
Director of Statistical Programmers – Peter Yribar
Finding the career that utilizes all of your talents can be unexpected with elements of luck.
Introductions to SAS and Clinical Trials
Pete was introduced to computer programming using Fortran and Cobol in school where he studied ORIS, Organization Research Information Systems. He also did some programming with IBM Assembler but he does not consider himself a programmer. His first position when he got out of school was working for a firm that did direct marketing. There were statisticians that used SAS so Pete started to learn more about statistical programming. He then began to take some SAS programming classes in the mid- eighties to apply this to his work.
Pete continued to work for five years in the marketing industry before he had an opportunity to work for Warner Lambert. This was a very different environment for him since organizationally the pharmaceutical company was much larger. When he first joined, SAS programmers had two distinct functions. They used SAS to handle the data management aspects of the operational clinical trials data and they also used SAS to perform analysis and reporting. This gave Pete a good introduction to clinical trials data from the point of data collection all the way through to the reports that were submitted to the FDA. Warner Lambert transitioned into using Oracle Clinical when that became available. At that point, the SAS programmer’s role gravitated more towards analysis and reporting since Oracle took over the data management domain.
While working with clinical trials data, Pete also learned many different aspects of SAS programming. He first learned to perform data manipulation and transformation with SAS data step. He then continued to learn more about programming with SCL which at the time stood for Screen Control Language. The language has been used in more diverse ways and now is known as SAS Component Language. Pete also automated many of the routines that were repetitive using the SAS macro language. This was performed on a variety of platforms including PC Windows, VMS and UNIX. He always felt that he was average as a programmer but thought that the job required greater attention to detail. He was able to come up with creative ideas and solutions, but felt that his tenacity for quality control was not sufficient to excel as a statistical analyst.
Managing SAS Programmers
After several years as a SAS programmer, he was promoted to manage programmers. The restructuring of the organization for SAS programmers to become more focused on analysis and reporting gave Pete an opportunity to transition into a management position. He could understand and appreciate what it took to perform the programming tasks so he was able to impart his empathy through management. It was not just managing what the team was doing that was important, but the management of each member’s expectations became very significant. SAS programmers analyzed the data but they also viewed developing SAS code as an expression of their creativity. The intellectual asset that was being developed into these programs was guarded with a certain sense of pride which was infused into the identity of the programmers. Pete suggests that if you were to ask four SAS programmers to come up with an analysis approach for the same problem, the result would be four different yet correct solutions. This presented challenges since each programmer would perceive their solution as being the best
Pete was fortunate to enter the industry at Warner Lambert during a period of tremendous growth. There were no other pharmaceutical companies in Michigan and therefore SAS programmers were well sought after. Even though Pete did not have experience with clinical trials experience when he first began, his SAS and general programming experience was enough to land him the position. He was a bit intimidated within the research environment at first. In his prior position within the marketing industry, team members with Master degrees were highly regarded and PhDs were rare. In the pharmaceutical industry, on the other hand, it was common to have many PhDs as part of the team. There was no formal training on clinical trials so Pete had to do a crash course while he was on the job. He was assigned to a project and given the protocol in a binder. He would then just read this and continue to refer to it as a resource for learning how clinical trials experiments were conducted. His process of learning was a trial and error approach. In retrospect, it may not have been the most efficient method but Pete felt that it was one that taught him a lot and the lessons learned have stayed with him. At the beginning, the simplest concept of a Case Report Form (CRF) eluded him. He worked on various projects but did not understand the bigger contextual concept of how the data was captured on these case report forms and then entered into the data base to be cleaned and then used for analysis. He had raised questions as to why and how but it took a couple of years before things became clearer.
Rather than working with the detailed level at each data point, Pete preferred to view a greater understanding of the process. He enjoyed finding ways of making the process work more efficiently at this level. He also had an eye for identifying talents and encouraging efficiency in the people which he hired.
Part of the challenge of managing SAS programmers was making the team feel that their contribution to the organization was valuable and appreciated. In this industry, the clinical scientist is revered and given much praise while everyone else is sometimes not appreciated as much. His challenge was to close this gap and show that each programmer was a valuable contributor.
A recruiter then contacted Pete about a position at Phyzer. Warner Lambert was considered to be a large organization when he joined but Phyzer proved to be even larger and had the bureaucracy to accompany it. He managed more people than ever before. Pete did not see too many differences between managing SAS programmers and managing other technical workers. Pete was in charge of managing the process for global stastical programmers. This process was then enforced at other divisions of Phyzer so it was very challenging trying to make this process work for all the different groups. It was a difficult position but Pete did learn a lot.
Pete was then once again contacted by a recruiter about a position at Genentech. Moving from Phyzer to Genentech was like night and day. Genentech was growing and was becoming larger but it functioned very differently as an organization and culture. One of the key cultural differences was that the underlying theme of the work that was being done was to save people’s lives. This aspect of the culture permeated all aspects of the organization at all levels. The philosophy united the organization and it no longer made SAS programmers feel that they were just a cog in the wheel. They belonged to something larger, a cause to which they could contribute. The individuals who were valued were no longer just the clinical scientists but rather there was a belief that everyone contributed. The job was not just to code in SAS, but it was more about the larger role of coming up with drugs that cure cancer and save people’s lives.
Pete thrived in this environment. He was able to bring his empathetic style of management to foster the talented team of SAS programmers with which he worked. Technical skills were essential but he stressed the importance of finding and developing people who wanted to work within the team with excellent communication skills. The programmer could be a good programmer, but if he or she could not work collaboratively within the group, then the talent would be considered worthless.
There was an element of luck and chance to arriving at the position where Pete is now. He was fortunate to be introduced to SAS programming in the market research group that he worked with out of school. His biggest break was when he landed a position at Waner Lambert. It was having the luck of being at the right place at the right time. That was a turning point in his career into the pharmaceutical industry. The many twists and turns which Pete took led him down a career path where he could fully utilize his talents.
All Roads Lead to SAS Programming
The diverse set of roads which these individuals took in their trip through SAS illustrates how the role as a clinical SAS programmer is evolving and new. The field is now becoming more defined as there are a greater body of experienced professionals within the field and the work is being better defined. The companies that employ SAS programmers are usually very large and tend to have their own culture. It therefore prevented inhibits a cross pollination of ideas and ways of working as it relates to SAS programming.
The industry is going through a period of change. With the standard organizations such as CDISC, growing popularity of professional groups such as PharmaSUG, all of the SAS conferences such as SAS Global Forum, and the burgeoning online community for SAS programmers, the position of a clinical SAS programmer is solidifying and is coming into its own.
The diversity of paths leading to the same road of becoming a SAS programmer is not a weakness but rather a strength. The dynamic nature of technologies both from the computer sciences and clinical sciences means that the professional who works in them has to adapt and evolve along with the technologies. Like in a biological organism, the diverse set of talents, backgrounds and personalities is supporting the development of a very vibrant and healthy set of professionals.
-----------------------
Horizontal Art placement
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- genesis chapter 1 questions and answers
- biology 101 chapter 1 quiz
- chapter 1 psychology test answers
- strategic management chapter 1 quiz
- psychology chapter 1 questions and answers
- cooper heron heward chapter 1 powerpoint
- chapter 1 psychology quiz
- chapter 1 what is psychology
- chapter 1 cooper heron heward
- medical terminology chapter 1 quiz
- holt physics chapter 1 test
- dod fmr volume 2a chapter 1 definitions